{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EB2YPQINTQJX3BNPMGSJNHGFBQ","short_pith_number":"pith:EB2YPQIN","schema_version":"1.0","canonical_sha256":"207587c10d9c137d85af61a4969cc50c048af5e258f9b793dc72fd9bda61607b","source":{"kind":"arxiv","id":"2412.04378","version":3},"attestation_state":"computed","paper":{"title":"VladVA: Discriminative Fine-tuning of LVLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Adrian Bulat, Alexandros Xenos, Anestis Zaganidis, Brais Martinez, Georgios Tzimiropoulos, Ioannis Maniadis Metaxas, Yassine Ouali","submitted_at":"2024-12-05T17:54:27Z","abstract_excerpt":"Contrastively-trained Vision-Language Models (VLMs) like CLIP have become the de facto approach for discriminative vision-language representation learning. However, these models have limited language understanding, often exhibiting a \"bag of words\" behavior. At the same time, Large Vision-Language Models (LVLMs), which combine vision encoders with LLMs, have been shown to be capable of detailed vision-language reasoning, yet their autoregressive nature renders them less suitable for discriminative tasks.\n  In this work, we propose to combine \"the best of both worlds\": a new training approach f"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2412.04378","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-12-05T17:54:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"118e18906f7df8c3bee8ec917f409343f973195af9c5c873d499da89a2759f6e","abstract_canon_sha256":"316848347a45262d38ade6820dbc288001d2c511dcfa594cc8f059076548afba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:00:25.646330Z","signature_b64":"xuZyGxc7gk0cbyjkocB551mr9hpysLGyKf44hONbjtGZEnhJOuztDOAiBF7tnDha8TMsnPI/KDKBEDH0pgG8DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"207587c10d9c137d85af61a4969cc50c048af5e258f9b793dc72fd9bda61607b","last_reissued_at":"2026-07-05T11:00:25.645778Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:00:25.645778Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VladVA: Discriminative Fine-tuning of LVLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Adrian Bulat, Alexandros Xenos, Anestis Zaganidis, Brais Martinez, Georgios Tzimiropoulos, Ioannis Maniadis Metaxas, Yassine Ouali","submitted_at":"2024-12-05T17:54:27Z","abstract_excerpt":"Contrastively-trained Vision-Language Models (VLMs) like CLIP have become the de facto approach for discriminative vision-language representation learning. However, these models have limited language understanding, often exhibiting a \"bag of words\" behavior. At the same time, Large Vision-Language Models (LVLMs), which combine vision encoders with LLMs, have been shown to be capable of detailed vision-language reasoning, yet their autoregressive nature renders them less suitable for discriminative tasks.\n  In this work, we propose to combine \"the best of both worlds\": a new training approach f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.04378","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2412.04378/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2412.04378","created_at":"2026-07-05T11:00:25.645848+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.04378v3","created_at":"2026-07-05T11:00:25.645848+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.04378","created_at":"2026-07-05T11:00:25.645848+00:00"},{"alias_kind":"pith_short_12","alias_value":"EB2YPQINTQJX","created_at":"2026-07-05T11:00:25.645848+00:00"},{"alias_kind":"pith_short_16","alias_value":"EB2YPQINTQJX3BNP","created_at":"2026-07-05T11:00:25.645848+00:00"},{"alias_kind":"pith_short_8","alias_value":"EB2YPQIN","created_at":"2026-07-05T11:00:25.645848+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.06144","citing_title":"CLaMR: Contextualized Late-Interaction for Multimodal Content Retrieval","ref_index":27,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EB2YPQINTQJX3BNPMGSJNHGFBQ","json":"https://pith.science/pith/EB2YPQINTQJX3BNPMGSJNHGFBQ.json","graph_json":"https://pith.science/api/pith-number/EB2YPQINTQJX3BNPMGSJNHGFBQ/graph.json","events_json":"https://pith.science/api/pith-number/EB2YPQINTQJX3BNPMGSJNHGFBQ/events.json","paper":"https://pith.science/paper/EB2YPQIN"},"agent_actions":{"view_html":"https://pith.science/pith/EB2YPQINTQJX3BNPMGSJNHGFBQ","download_json":"https://pith.science/pith/EB2YPQINTQJX3BNPMGSJNHGFBQ.json","view_paper":"https://pith.science/paper/EB2YPQIN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.04378&json=true","fetch_graph":"https://pith.science/api/pith-number/EB2YPQINTQJX3BNPMGSJNHGFBQ/graph.json","fetch_events":"https://pith.science/api/pith-number/EB2YPQINTQJX3BNPMGSJNHGFBQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EB2YPQINTQJX3BNPMGSJNHGFBQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EB2YPQINTQJX3BNPMGSJNHGFBQ/action/storage_attestation","attest_author":"https://pith.science/pith/EB2YPQINTQJX3BNPMGSJNHGFBQ/action/author_attestation","sign_citation":"https://pith.science/pith/EB2YPQINTQJX3BNPMGSJNHGFBQ/action/citation_signature","submit_replication":"https://pith.science/pith/EB2YPQINTQJX3BNPMGSJNHGFBQ/action/replication_record"}},"created_at":"2026-07-05T11:00:25.645848+00:00","updated_at":"2026-07-05T11:00:25.645848+00:00"}